The Research Of The Multifractal Characteristics And Risk Of Energy Market Based On Time-window

Price volatility of crude oil and natural gas are closely related to people's life and the economic development of countries.Especially in recent years,soaring and plummeting of crude oil price have a significant impact on the macroeconomic of the different countries.The international energy agency has estimated that 80% of oil consumption and 50% of natural gas consumption in China will be dependent on imports by 2035.Therefore,it is very important for countries and investors to explore the rule of their prices and returns' volatility.In recent years,more and more scholars have used the Fractal Theory to detect the complexity and risk of the financial market.However,the present researches mainly focus on the study of the characteristics of the whole time series,which can not reflect the local characteristics as well as the size of the risk depending on the time.Based on this point,this paper proposes the time-window fractal model to study the market characteristics with the tiny change over time.And the advantage of this model is that it not only can analyze the local fractal features but also can reflect the influence of extreme risk on the fractal features.The main conclusions are as follows:(1)Crude oil market and natural gas market are not efficient markets,and the price series and return series of the two markets have multifractal characteristic.The main factors of multifractal characteristic come from the long-range correlation of the sequence fluctuations and fat-tail characteristics of probability distribution.Crude oil market is an emerging market,while the natural gas market is a developed markets.Namely,the risk of crude oil market is higher than that of the natural gas market.The natural gas market affects crude oil market when the returns fluctuate in small ranges,and the crude oil market affects the natural gas market when the returns fluctuate in large ranges.(2)Using accumulating time window fractal model to analyze these two return series,we can found that the variations of the mulitfracal strength tend to be gentle when the time series contains five years of return fluctuation information.Hereafter,the growth of the series will not cause the sharp fluctuation of the variations of the mulitfractality.This shows that the time series need at least 5 years of the data to reflect the long time information of the markets.Unlike the natural gas market,the plummeting of the crude oil price made the long-term investors lost confidence to the market in the second half of 2014.They stopped the market operation or transaction based on short-term investment information,so the long-term market information has changed,and the stability of the fractal characteristics of the market was broken.(3)Using sliding time window fractal model to analyze these two return series,we found that the short-term characteristics of the crude oil market were affected by weekly volatility of the crude oil return.Crude oil price fell below $60 per barrel for the first time caused the strongest panic,the strongest flock of effects and the strongest multifractal strength.The fractal characteristics of the crude oil market and the natural gas market present an opposite change over time,and the risk of the crude oil market is greater than that of the natural gas market.